submission_id: trace2333-mistral-align-_5060_v3
developer_uid: Trace2333
best_of: 8
celo_rating: 1254.68
display_name: trace2333-mistral-align-_5060_v3
family_friendly_score: 0.0
formatter: {'memory_template': "{bot_name}'s Persona: {memory}\n####\n", 'prompt_template': '{prompt}\n<START>\n', 'bot_template': '{bot_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '{bot_name}:', 'truncate_by_message': False}
generation_params: {'temperature': 0.9, 'top_p': 1.0, 'min_p': 0.05, 'top_k': 80, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '</s>', '###'], 'max_input_tokens': 512, 'best_of': 8, 'max_output_tokens': 64}
gpu_counts: {'NVIDIA RTX A5000': 1}
is_internal_developer: False
language_model: Trace2333/mistral_align_namo_3982
latencies: [{'batch_size': 1, 'throughput': 0.69193837225577, 'latency_mean': 1.4451543378829956, 'latency_p50': 1.4565082788467407, 'latency_p90': 1.6050782203674316}, {'batch_size': 3, 'throughput': 1.3269517863713494, 'latency_mean': 2.252337239980698, 'latency_p50': 2.246236205101013, 'latency_p90': 2.496141982078552}, {'batch_size': 5, 'throughput': 1.5640747785590146, 'latency_mean': 3.1804043531417845, 'latency_p50': 3.18123459815979, 'latency_p90': 3.5621174335479737}, {'batch_size': 6, 'throughput': 1.5816621757667753, 'latency_mean': 3.768118335008621, 'latency_p50': 3.762748122215271, 'latency_p90': 4.2527915954589846}, {'batch_size': 8, 'throughput': 1.5896844039815494, 'latency_mean': 5.003578040599823, 'latency_p50': 5.008600473403931, 'latency_p90': 5.659723377227783}, {'batch_size': 10, 'throughput': 1.5248741389252851, 'latency_mean': 6.520109952688217, 'latency_p50': 6.586364388465881, 'latency_p90': 7.38736629486084}]
max_input_tokens: 512
max_output_tokens: 64
model_architecture: MistralForCausalLM
model_group: Trace2333/mistral_align_
model_name: trace2333-mistral-align-_5060_v3
model_num_parameters: 12772070400.0
model_repo: Trace2333/mistral_align_namo_3982
model_size: 13B
num_battles: 13267
num_wins: 6889
ranking_group: single
status: torndown
submission_type: basic
throughput_3p7s: 1.59
timestamp: 2024-09-07T07:44:39+00:00
us_pacific_date: 2024-09-07
win_ratio: 0.5192583100927113
Download Preference Data
Resubmit model
Shutdown handler not registered because Python interpreter is not running in the main thread
run pipeline %s
run pipeline stage %s
Running pipeline stage MKMLizer
Starting job with name trace2333-mistral-align-5060-v3-mkmlizer
Waiting for job on trace2333-mistral-align-5060-v3-mkmlizer to finish
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
trace2333-mistral-align-5060-v3-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
trace2333-mistral-align-5060-v3-mkmlizer: ║ _____ __ __ ║
trace2333-mistral-align-5060-v3-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
trace2333-mistral-align-5060-v3-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
trace2333-mistral-align-5060-v3-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
trace2333-mistral-align-5060-v3-mkmlizer: ║ /___/ ║
trace2333-mistral-align-5060-v3-mkmlizer: ║ ║
trace2333-mistral-align-5060-v3-mkmlizer: ║ Version: 0.10.1 ║
trace2333-mistral-align-5060-v3-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
trace2333-mistral-align-5060-v3-mkmlizer: ║ https://mk1.ai ║
trace2333-mistral-align-5060-v3-mkmlizer: ║ ║
trace2333-mistral-align-5060-v3-mkmlizer: ║ The license key for the current software has been verified as ║
trace2333-mistral-align-5060-v3-mkmlizer: ║ belonging to: ║
trace2333-mistral-align-5060-v3-mkmlizer: ║ ║
trace2333-mistral-align-5060-v3-mkmlizer: ║ Chai Research Corp. ║
trace2333-mistral-align-5060-v3-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
trace2333-mistral-align-5060-v3-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
trace2333-mistral-align-5060-v3-mkmlizer: ║ ║
trace2333-mistral-align-5060-v3-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
trace2333-mistral-align-5060-v3-mkmlizer: Downloaded to shared memory in 29.992s
trace2333-mistral-align-5060-v3-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpqt9aiez0, device:0
trace2333-mistral-align-5060-v3-mkmlizer: Saving flywheel model at /dev/shm/model_cache
trace2333-mistral-align-5060-v3-mkmlizer: quantized model in 36.542s
trace2333-mistral-align-5060-v3-mkmlizer: Processed model Trace2333/mistral_align_namo_3982 in 66.535s
trace2333-mistral-align-5060-v3-mkmlizer: creating bucket guanaco-mkml-models
trace2333-mistral-align-5060-v3-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
trace2333-mistral-align-5060-v3-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/trace2333-mistral-align-5060-v3
trace2333-mistral-align-5060-v3-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/trace2333-mistral-align-5060-v3/config.json
trace2333-mistral-align-5060-v3-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/trace2333-mistral-align-5060-v3/special_tokens_map.json
trace2333-mistral-align-5060-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/trace2333-mistral-align-5060-v3/tokenizer_config.json
trace2333-mistral-align-5060-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/trace2333-mistral-align-5060-v3/tokenizer.json
trace2333-mistral-align-5060-v3-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/trace2333-mistral-align-5060-v3/flywheel_model.0.safetensors
trace2333-mistral-align-5060-v3-mkmlizer: Loading 0: 0%| | 0/363 [00:00<?, ?it/s] Loading 0: 2%|▏ | 7/363 [00:00<00:06, 52.41it/s] Loading 0: 6%|▌ | 22/363 [00:00<00:04, 83.25it/s] Loading 0: 9%|▊ | 31/363 [00:00<00:04, 81.98it/s] Loading 0: 11%|█ | 40/363 [00:00<00:03, 82.59it/s] Loading 0: 13%|█▎ | 49/363 [00:00<00:03, 79.22it/s] Loading 0: 16%|█▌ | 58/363 [00:00<00:03, 77.83it/s] Loading 0: 18%|█▊ | 66/363 [00:01<00:15, 19.48it/s] Loading 0: 21%|██ | 76/363 [00:02<00:11, 25.88it/s] Loading 0: 23%|██▎ | 85/363 [00:02<00:08, 32.77it/s] Loading 0: 26%|██▌ | 94/363 [00:02<00:06, 39.58it/s] Loading 0: 28%|██▊ | 103/363 [00:02<00:05, 47.43it/s] Loading 0: 31%|███ | 112/363 [00:02<00:04, 54.40it/s] Loading 0: 33%|███▎ | 121/363 [00:02<00:04, 60.43it/s] Loading 0: 36%|███▌ | 130/363 [00:02<00:03, 63.85it/s] Loading 0: 38%|███▊ | 139/363 [00:02<00:03, 64.67it/s] Loading 0: 40%|████ | 147/363 [00:03<00:10, 19.66it/s] Loading 0: 42%|████▏ | 153/363 [00:04<00:09, 22.31it/s] Loading 0: 44%|████▍ | 160/363 [00:04<00:07, 27.16it/s] Loading 0: 47%|████▋ | 169/363 [00:04<00:05, 35.50it/s] Loading 0: 49%|████▉ | 178/363 [00:04<00:04, 43.06it/s] Loading 0: 52%|█████▏ | 187/363 [00:04<00:03, 49.65it/s] Loading 0: 54%|█████▍ | 196/363 [00:04<00:02, 57.53it/s] Loading 0: 56%|█████▋ | 205/363 [00:04<00:02, 64.19it/s] Loading 0: 59%|█████▉ | 214/363 [00:04<00:02, 69.71it/s] Loading 0: 61%|██████▏ | 223/363 [00:06<00:06, 20.09it/s] Loading 0: 64%|██████▍ | 232/363 [00:06<00:05, 26.06it/s] Loading 0: 66%|██████▋ | 241/363 [00:06<00:03, 32.08it/s] Loading 0: 69%|██████▉ | 250/363 [00:06<00:02, 38.32it/s] Loading 0: 71%|███████▏ | 259/363 [00:06<00:02, 46.20it/s] Loading 0: 74%|███████▍ | 268/363 [00:06<00:01, 52.74it/s] Loading 0: 76%|███████▋ | 277/363 [00:06<00:01, 59.18it/s] Loading 0: 79%|███████▉ | 286/363 [00:06<00:01, 65.30it/s] Loading 0: 82%|████████▏ | 296/363 [00:06<00:00, 73.14it/s] Loading 0: 84%|████████▍ | 305/363 [00:08<00:02, 20.85it/s] Loading 0: 86%|████████▌ | 313/363 [00:08<00:01, 25.53it/s] Loading 0: 89%|████████▉ | 323/363 [00:08<00:01, 33.65it/s] Loading 0: 91%|█████████ | 331/363 [00:08<00:00, 39.47it/s] Loading 0: 94%|█████████▎| 340/363 [00:08<00:00, 45.77it/s] Loading 0: 96%|█████████▌| 349/363 [00:08<00:00, 51.53it/s] Loading 0: 99%|█████████▊| 358/363 [00:08<00:00, 59.10it/s]
Job trace2333-mistral-align-5060-v3-mkmlizer completed after 147.5s with status: succeeded
Stopping job with name trace2333-mistral-align-5060-v3-mkmlizer
Pipeline stage MKMLizer completed in 148.89s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.15s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service trace2333-mistral-align-5060-v3
Waiting for inference service trace2333-mistral-align-5060-v3 to be ready
Failed to get response for submission blend_lobuf_2024-08-22: ('http://zonemercy-taurus-edit-v1-1e5-v3-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:58842->127.0.0.1:8080: read: connection reset by peer\n')
Inference service trace2333-mistral-align-5060-v3 ready after 150.7334201335907s
Pipeline stage MKMLDeployer completed in 151.08s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.3135838508605957s
Received healthy response to inference request in 1.9595134258270264s
Received healthy response to inference request in 2.3065500259399414s
Received healthy response to inference request in 3.0247418880462646s
Received healthy response to inference request in 2.1177732944488525s
5 requests
0 failed requests
5th percentile: 1.9911653995513916
10th percentile: 2.022817373275757
20th percentile: 2.0861213207244873
30th percentile: 2.1555286407470704
40th percentile: 2.2310393333435057
50th percentile: 2.3065500259399414
60th percentile: 2.3093635559082033
70th percentile: 2.3121770858764648
80th percentile: 2.45581545829773
90th percentile: 2.7402786731719972
95th percentile: 2.8825102806091305
99th percentile: 2.996295566558838
mean time: 2.344432497024536
Pipeline stage StressChecker completed in 12.52s
run pipeline stage %s
Running pipeline stage TriggerMKMLProfilingPipeline
run_pipeline:run_in_cloud %s
starting trigger_guanaco_pipeline args=%s
Pipeline stage TriggerMKMLProfilingPipeline completed in 4.48s
Shutdown handler de-registered
trace2333-mistral-align-_5060_v3 status is now deployed due to DeploymentManager action
Shutdown handler registered
run pipeline %s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Skipping teardown as no inference service was successfully deployed
Pipeline stage MKMLProfilerDeleter completed in 0.11s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.09s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service trace2333-mistral-align-5060-v3-profiler
Waiting for inference service trace2333-mistral-align-5060-v3-profiler to be ready
Inference service trace2333-mistral-align-5060-v3-profiler ready after 150.43066120147705s
Pipeline stage MKMLProfilerDeployer completed in 151.61s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/trace2333-mistral-ala25507e6ede684dafad2c4ba2a9dbcd0-deplokxl8f:/code/chaiverse_profiler_1725695586 --namespace tenant-chaiml-guanaco
kubectl exec -it trace2333-mistral-ala25507e6ede684dafad2c4ba2a9dbcd0-deplokxl8f --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1725695586 && python profiles.py profile --best_of_n 8 --auto_batch 5 --batches 1,5,10,15,20,25,30,35,40,45,50,55,60,65,70,75,80,85,90,95,100,105,110,115,120,125,130,135,140,145,150,155,160,165,170,175,180,185,190,195 --samples 200 --input_tokens 512 --output_tokens 64 --summary /code/chaiverse_profiler_1725695586/summary.json'
kubectl exec -it trace2333-mistral-ala25507e6ede684dafad2c4ba2a9dbcd0-deplokxl8f --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1725695586/summary.json'
Pipeline stage MKMLProfilerRunner completed in 956.11s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service trace2333-mistral-align-5060-v3-profiler is running
Tearing down inference service trace2333-mistral-align-5060-v3-profiler
Service trace2333-mistral-align-5060-v3-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.76s
Shutdown handler de-registered
trace2333-mistral-align-_5060_v3 status is now inactive due to auto deactivation removed underperforming models
trace2333-mistral-align-_5060_v3 status is now torndown due to DeploymentManager action
trace2333-mistral-align-_5060_v3 status is now torndown due to DeploymentManager action